Written by Jakub Rusinowski · Last updated August 15, 2026
Model library → Nemotron 3 Super → Nemotron 3 Super 120B-A12B
120B total / 12B active MoE. The low active-parameter count means throughput closer to a 12B model than a 120B one, but the full weight set still has to be resident — roughly 65 GB at Q4_K_M, so dual RTX 3090/4090 or a single 80GB datacenter card. Q8_0 and BF16 tags are published for workstations with more headroom. NVIDIA releases open training data and recipes with this family, which is unusual among open-weight releases.
Nemotron 3 Super 120B-A12B needs about 73 GB of VRAM at Q4_K_M — quantized weights plus framework overhead, before any KV cache. On Apple Silicon that figure comes out of unified memory.
| Parameters | 120 Billion (12B active) |
| Context window | 128,000 |
| Architecture | Mixture-of-Experts (reasoning-trace) |
| Provider | NVIDIA |
| Licence | NVIDIA Open Model License |
| Specified at | Q4_K_M |
| System RAM | 128 GB |
| Record updated | 2026-08-15 |
NVIDIA Open Model License — commercial use permitted. Weights are downloadable and commercial use is permitted, subject to the licence’s acceptable-use terms.
Modelled on a reference NVIDIA RTX 4090 (24 GB), with no KV cache (this record has no published architecture). Speed figures are ESTIMATES from the memory-bandwidth roofline described on the methodology page, not benchmarks we ran — rows marked measured come from published or reader-submitted runs. VRAM here includes the KV cache, so it reads higher than the headline figure above, which does not.
| Quant | Weights | VRAM needed | Est. speed | Fit on 24 GB |
|---|---|---|---|---|
| Q2_K | 39.4 GB | 40.2 GB | ~16 tok/s (est.) | Offloads to system RAM (slow) |
| Q3_K_M | 51.2 GB | 52.0 GB | ~14 tok/s (est.) | Offloads to system RAM (slow) |
| Q4_K_M | 72.5 GB | 73.3 GB | — | Won't fit |
| Q5_K_M | 85.0 GB | 85.8 GB | — | Won't fit |
| Q6_K | 98.4 GB | 99.2 GB | — | Won't fit |
| Q8_0 | 127.5 GB | 128.3 GB | — | Won't fit |
| F16 | 240.0 GB | 240.8 GB | — | Won't fit |
Want the memory numbers alone, at every quantization level and your own context length? Use the Nemotron 3 Super 120B-A12B VRAM calculator.
or compare on Vast.ai from $0.77/hr (typical low · varies)
As an Amazon Associate we earn from qualifying purchases. Cloud GPU links are referral links — we may earn a commission at no extra cost to you.
The cheapest catalogued GPU that runs Nemotron 3 Super 120B-A12B is the AMD Ryzen AI Max+ 395 (96 GB).
Install Ollama, then run:
ollama run nemotron-3-super:120b-a12bWeights on Hugging Face: nvidia/Nemotron-3-Super-120B-A12B.
Best for: agentic tasks, reasoning, enterprise, on premise, multi agent.
← All Nemotron 3 Super models | VRAM calculator | Check your own hardware